Abstract
To date, top-performing optical flow estimation methods only take pairs of consecutive frames into account. While elegant and appealing, the idea of using more than two frames has not yet produced state-of-the-art results. We present a simple, yet effective fusion approach for multi-frame optical flow that benefits from longer-term temporal cues. Our method first warps the optical flow from previous frames to the current, thereby yielding multiple plausible estimates. It then fuses the complementary information carried by these estimates into a new optical flow field. At the time of writing, our method ranks first among published results in the MPI Sintel and KITTI 2015 benchmarks.
| Original language | English |
|---|---|
| Title of host publication | Computer Vision – ECCV 2018 Workshops, Proceedings |
| Editors | Laura Leal-Taixé, Stefan Roth |
| Publisher | Springer Verlag |
| Pages | 706-710 |
| Number of pages | 5 |
| ISBN (Print) | 9783030110239 |
| DOIs | |
| Publication status | Published - 2019 Jan 1 |
| Event | 15th European Conference on Computer Vision, ECCV 2018 - Munich, Germany Duration: 2018 Sep 8 → 2018 Sep 14 |
Publication series
| Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
|---|---|
| Volume | 11134 LNCS |
| ISSN (Print) | 0302-9743 |
| ISSN (Electronic) | 1611-3349 |
Other
| Other | 15th European Conference on Computer Vision, ECCV 2018 |
|---|---|
| Country | Germany |
| City | Munich |
| Period | 18/9/8 → 18/9/14 |
Fingerprint
All Science Journal Classification (ASJC) codes
- Theoretical Computer Science
- Computer Science(all)
Cite this
}
A simple and effective fusion approach for multi-frame optical flow estimation. / Ren, Zhile; Gallo, Orazio; Sun, Deqing; Yang, Ming Hsuan; Sudderth, Erik B.; Kautz, Jan.
Computer Vision – ECCV 2018 Workshops, Proceedings. ed. / Laura Leal-Taixé; Stefan Roth. Springer Verlag, 2019. p. 706-710 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11134 LNCS).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
TY - GEN
T1 - A simple and effective fusion approach for multi-frame optical flow estimation
AU - Ren, Zhile
AU - Gallo, Orazio
AU - Sun, Deqing
AU - Yang, Ming Hsuan
AU - Sudderth, Erik B.
AU - Kautz, Jan
PY - 2019/1/1
Y1 - 2019/1/1
N2 - To date, top-performing optical flow estimation methods only take pairs of consecutive frames into account. While elegant and appealing, the idea of using more than two frames has not yet produced state-of-the-art results. We present a simple, yet effective fusion approach for multi-frame optical flow that benefits from longer-term temporal cues. Our method first warps the optical flow from previous frames to the current, thereby yielding multiple plausible estimates. It then fuses the complementary information carried by these estimates into a new optical flow field. At the time of writing, our method ranks first among published results in the MPI Sintel and KITTI 2015 benchmarks.
AB - To date, top-performing optical flow estimation methods only take pairs of consecutive frames into account. While elegant and appealing, the idea of using more than two frames has not yet produced state-of-the-art results. We present a simple, yet effective fusion approach for multi-frame optical flow that benefits from longer-term temporal cues. Our method first warps the optical flow from previous frames to the current, thereby yielding multiple plausible estimates. It then fuses the complementary information carried by these estimates into a new optical flow field. At the time of writing, our method ranks first among published results in the MPI Sintel and KITTI 2015 benchmarks.
UR - http://www.scopus.com/inward/record.url?scp=85061740901&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85061740901&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-11024-6_53
DO - 10.1007/978-3-030-11024-6_53
M3 - Conference contribution
AN - SCOPUS:85061740901
SN - 9783030110239
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 706
EP - 710
BT - Computer Vision – ECCV 2018 Workshops, Proceedings
A2 - Leal-Taixé, Laura
A2 - Roth, Stefan
PB - Springer Verlag
ER -